| Name | nirtorch JSON |
| Version |
1.0
JSON |
| download |
| home_page | |
| Summary | Neuromorphic Intermediate Representation |
| upload_time | 2023-12-06 14:32:04 |
| maintainer | |
| docs_url | None |
| author | |
| requires_python | |
| license | |
| keywords |
neuromorphic
intermediate
representation
|
| VCS |
|
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# NIRTorch
PyTorch helpers for the [Neuromorphic Intermediate Representation (NIR)](https://github.com/neuromorphs/nir).
This is a no frills python package to enable torch based libraries to translate to NIR.
## Installation
```shell
pip install nirtorch
```
## Usage
NIRTorch is typically only interfaced by library/hardwarae developers.
NIRTorch provides the `extract_nir_graph` function that takes as input a `torch.nn.Module` and a means to map Torch modules into NIR nodes.
An NIR node is an element in the NIR compute graph, corresponding to neuromorphic ODEs.
Here is an example from the [Norse](https://github.com/norse/norse) library:
```python
def _extract_norse_module(module: torch.nn.Module) -> Optional[nir.NIRNode]:
if isinstance(module, LIFBoxCell):
return nir.LIF(
tau=module.p.tau_mem_inv,
v_th=module.p.v_th,
v_leak=module.p.v_leak,
r=torch.ones_like(module.p.v_leak),
)
elif isinstance(module, torch.nn.Linear):
return nir.Linear(module.weight, module.bias)
elif ...
return None
def to_nir(
module: torch.nn.Module, sample_data: torch.Tensor, model_name: str = "norse"
) -> nir.NIRNode:
return extract_nir_graph(
module, _extract_norse_module, sample_data, model_name=model_name
)
```
## Acknowledgements
If you use NIR torch in your work, please cite the [following Zenodo reference](https://zenodo.org/record/8105042)
```
@software{nir2023,
author = {Abreu, Steven and
Bauer, Felix and
Eshraghian, Jason and
Jobst, Matthias and
Lenz, Gregor and
Pedersen, Jens Egholm and
Sheik, Sadique},
title = {Neuromorphic Intermediate Representation},
month = jul,
year = 2023,
publisher = {Zenodo},
version = {0.0.1},
doi = {10.5281/zenodo.8105042},
url = {https://doi.org/10.5281/zenodo.8105042}
}
```
## For developers
If you want to make sure that your code is linted correctly on your local machine, use [pre-commit](https://pre-commit.com/) to automatically perform checks before every git commit. To use it, first install the package in your environment
```
pip install pre-commit
```
and then install the pre-commit hooks that are listed in the root of this repository
```
pre-commit install
```
Next time you commit some changes, all the checks will be run!
Raw data
{
"_id": null,
"home_page": "",
"name": "nirtorch",
"maintainer": "",
"docs_url": null,
"requires_python": "",
"maintainer_email": "",
"keywords": "neuromorphic,intermediate,representation",
"author": "",
"author_email": "Steven Abreu <s.abreu@rug.nl>, Felix Bauer <felix.bauer@synsen.ai>, Jason Eshraghian <jeshragh@ucsc.edu>, Matthias Jobst <matthias.jobst2@tu-dresden.de>, Gregor Lenz <mail@lenzgregor.com>, Jens Egholm Pedersen <jens@jepedersen.dk>, Sadique Sheik <sadique.sheik@synsense.ai>",
"download_url": "https://files.pythonhosted.org/packages/fd/1b/27cb50acf5a0afcca55f238e32ea2a1ed9c5b58225346170682cd58fcfd6/nirtorch-1.0.tar.gz",
"platform": null,
"description": "# NIRTorch\n\nPyTorch helpers for the [Neuromorphic Intermediate Representation (NIR)](https://github.com/neuromorphs/nir).\nThis is a no frills python package to enable torch based libraries to translate to NIR.\n\n## Installation\n```shell\npip install nirtorch\n```\n\n## Usage\n\nNIRTorch is typically only interfaced by library/hardwarae developers.\nNIRTorch provides the `extract_nir_graph` function that takes as input a `torch.nn.Module` and a means to map Torch modules into NIR nodes.\nAn NIR node is an element in the NIR compute graph, corresponding to neuromorphic ODEs.\n\nHere is an example from the [Norse](https://github.com/norse/norse) library:\n\n```python\ndef _extract_norse_module(module: torch.nn.Module) -> Optional[nir.NIRNode]:\n if isinstance(module, LIFBoxCell):\n return nir.LIF(\n tau=module.p.tau_mem_inv,\n v_th=module.p.v_th,\n v_leak=module.p.v_leak,\n r=torch.ones_like(module.p.v_leak),\n )\n elif isinstance(module, torch.nn.Linear):\n return nir.Linear(module.weight, module.bias)\n elif ...\n\n return None\n\ndef to_nir(\n module: torch.nn.Module, sample_data: torch.Tensor, model_name: str = \"norse\"\n) -> nir.NIRNode:\n return extract_nir_graph(\n module, _extract_norse_module, sample_data, model_name=model_name\n )\n```\n\n## Acknowledgements\nIf you use NIR torch in your work, please cite the [following Zenodo reference](https://zenodo.org/record/8105042)\n\n```\n@software{nir2023,\n author = {Abreu, Steven and\n Bauer, Felix and\n Eshraghian, Jason and\n Jobst, Matthias and\n Lenz, Gregor and\n Pedersen, Jens Egholm and\n Sheik, Sadique},\n title = {Neuromorphic Intermediate Representation},\n month = jul,\n year = 2023,\n publisher = {Zenodo},\n version = {0.0.1},\n doi = {10.5281/zenodo.8105042},\n url = {https://doi.org/10.5281/zenodo.8105042}\n}\n```\n\n## For developers\nIf you want to make sure that your code is linted correctly on your local machine, use [pre-commit](https://pre-commit.com/) to automatically perform checks before every git commit. To use it, first install the package in your environment\n```\npip install pre-commit\n```\nand then install the pre-commit hooks that are listed in the root of this repository\n```\npre-commit install\n```\nNext time you commit some changes, all the checks will be run!\n",
"bugtrack_url": null,
"license": "",
"summary": "Neuromorphic Intermediate Representation",
"version": "1.0",
"project_urls": null,
"split_keywords": [
"neuromorphic",
"intermediate",
"representation"
],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "cd7492cc684fd83636b072318693676877af0d80c4e136067237f147f9a18d6f",
"md5": "3f352af0eee0e737f73057087161a15f",
"sha256": "72f4898a18ceb91d5a9b66515db2bb4a5893a8d11a77b48d9bbe058228995b02"
},
"downloads": -1,
"filename": "nirtorch-1.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "3f352af0eee0e737f73057087161a15f",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": null,
"size": 13313,
"upload_time": "2023-12-06T14:32:03",
"upload_time_iso_8601": "2023-12-06T14:32:03.016806Z",
"url": "https://files.pythonhosted.org/packages/cd/74/92cc684fd83636b072318693676877af0d80c4e136067237f147f9a18d6f/nirtorch-1.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "fd1b27cb50acf5a0afcca55f238e32ea2a1ed9c5b58225346170682cd58fcfd6",
"md5": "6887742a99149ad7b5c5644f8fefa571",
"sha256": "a0a7c485c0cb523a257a68b881c83d3710a471ad896188879a1bb722d47db2bf"
},
"downloads": -1,
"filename": "nirtorch-1.0.tar.gz",
"has_sig": false,
"md5_digest": "6887742a99149ad7b5c5644f8fefa571",
"packagetype": "sdist",
"python_version": "source",
"requires_python": null,
"size": 18870,
"upload_time": "2023-12-06T14:32:04",
"upload_time_iso_8601": "2023-12-06T14:32:04.023389Z",
"url": "https://files.pythonhosted.org/packages/fd/1b/27cb50acf5a0afcca55f238e32ea2a1ed9c5b58225346170682cd58fcfd6/nirtorch-1.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-12-06 14:32:04",
"github": false,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"lcname": "nirtorch"
}
